Hi Guillaume, 

 that's fantastic, thank you very much! I think I'm fine with otbConvert 
and ExtractROI, then. =)

Regards,
Robin



Am Freitag, 5. September 2014 10:16:39 UTC+2 schrieb Guillaume Pasero:
>
>  Hi Robin,
>
> The application Convert doesn't not allow to do band re-composition. You 
> have to use the application ExtractROI for that. If you don't specify the 
> ROI region, the whole image is used : 
>
> otbcli_ExtractROI -in your_multispectral_image.tif
>                   -out output_3bands_image.tif
>                   -cl Channel5 Channel3 Channel2
>
>
> The channel numbers have to be prefixed with "Channel". First channel is 
> Channel1.
> If your input image has pixel encoded on 16bit unsigned integers, you can 
> also add "uint16" after the output file name, so that the image produced 
> uses this pixel type. The default pixel type is float (takes more space).
>
> Regards,
> Guillaume
>
> Le 04/09/2014 20:53, Robin Holler a écrit :
>  
> One more thx Guillaume! =)
>
>  I understand that my image is edited in tiles and I have to calculate and 
> use absolute min/max values to be sure, each tile is processed the same 
> way. Thx for that! I'll try soon, but I first have to make other things 
> running and to do that faster, I'd like to know, if otbConvert is able to 
> extract three selected bands only while converting format? 
>
>  I'd like to give three integers (e.g. 5,3,2) as arguments and having 
> converted a multispectral image TIFF to a 3-channel RGB PNG, for instance. 
> Is that possible?
>
> Thx again,
> Robin
>
>
>
> Am Mittwoch, 3. September 2014 14:21:13 UTC+2 schrieb Guillaume Pasero: 
>>
>>  Hi Robin,
>>
>> There is a cleaner way to do it. 
>>
>> In your example, the input minimum and maximum are not given to the 
>> VectorRescaler, so it has to estimate them on each buffered region (i.e. 
>> each stream block). This is why you get a different rescaling  for each 
>> block (some block may also have black borders, which gives a "wrong" 
>> minimum value of 0).
>>
>> What you can do is :
>> - estimate the min / max values in your input image using an 
>> otb::StreamingStatisticsVectorImageFilter.
>> - set the input min/max values in the rescale filter.
>> - disable the automatic computation of input min/max
>>
>> You can have a look to the application Convert 
>> (Applications/Utils/otbConvert.cxx), which does nearly the same thing.
>>
>> Regards,
>> Guillaume
>>
>>
>> Le 03/09/2014 13:33, Robin Holler a écrit :
>>  
>> I'm as glad as proud having resolved this by myself. Don't know if it's 
>> the finest solution, but for now it works. The problem was writing output 
>> through streaming TIFF in tiles ... some researching brought 
>> http://wiki.orfeo-toolbox.org/index.php/Writing_large_images and 
>> http://www.orfeo-toolbox.org/doxygen-current/otbImageFileWriter_8h_source.html
>>  
>> ... So I did some experiments and since export GDAL_CACHEMAX=256 didn't 
>> have any effect I finally ended with changing to stripped Streaming but 
>> without dividing in lines or tiles:
>>
>>          Writer->SetAutomaticStrippedStreaming();
>>         Writer->SetNumberOfLinesStrippedStreaming(0);
>>  
>> Works for now, but maybe somebody once will come up with a better (more 
>> experienced) solution.
>>
>> Regards,
>> Robin
>>  -- 
>> -- 
>> Check the OTB FAQ at
>> http://www.orfeo-toolbox.org/FAQ.html
>>  
>> You received this message because you are subscribed to the Google
>> Groups "otb-users" group.
>> To post to this group, send email to [email protected]
>> To unsubscribe from this group, send email to
>> [email protected]
>> For more options, visit this group at
>> http://groups.google.com/group/otb-users?hl=en
>> --- 
>> You received this message because you are subscribed to the Google Groups 
>> "otb-users" group.
>> To unsubscribe from this group and stop receiving emails from it, send an 
>> email to [email protected].
>> For more options, visit https://groups.google.com/d/optout.
>>
>>
>>
>> -- 
>>   <http://www.c-s.fr>  *Guillaume PASERO*
>> Ingénieur d'études et développement
>>  *Business Unit E-SPACE & Geo Information* 
>> <https://thor.si.c-s.fr/blogs/cs-blogs-business/>* - Département 
>> APPLICATIONS*
>>
>> *CS Systèmes d'Information*
>> Parc de la Grande Plaine - 5, Rue Brindejonc des Moulinais - BP 15872
>> 31506 Toulouse Cedex 05 - FRANCE
>> +33 561 17 64 21 - [email protected]   
>>  
>  -- 
> -- 
> Check the OTB FAQ at
> http://www.orfeo-toolbox.org/FAQ.html
>  
> You received this message because you are subscribed to the Google
> Groups "otb-users" group.
> To post to this group, send email to [email protected] 
> <javascript:>
> To unsubscribe from this group, send email to
> [email protected] <javascript:>
> For more options, visit this group at
> http://groups.google.com/group/otb-users?hl=en
> --- 
> You received this message because you are subscribed to the Google Groups 
> "otb-users" group.
> To unsubscribe from this group and stop receiving emails from it, send an 
> email to [email protected] <javascript:>.
> For more options, visit https://groups.google.com/d/optout.
>
>
>
> -- 
>   <http://www.c-s.fr>  *Guillaume PASERO*
> Ingénieur d'études et développement
>  *Business Unit E-SPACE & Geo Information* 
> <https://thor.si.c-s.fr/blogs/cs-blogs-business/>* - Département 
> APPLICATIONS*
>
> *CS Systèmes d'Information*
> Parc de la Grande Plaine - 5, Rue Brindejonc des Moulinais - BP 15872
> 31506 Toulouse Cedex 05 - FRANCE
> +33 561 17 64 21 - [email protected] <javascript:>   
>  

-- 
-- 
Check the OTB FAQ at
http://www.orfeo-toolbox.org/FAQ.html

You received this message because you are subscribed to the Google
Groups "otb-users" group.
To post to this group, send email to [email protected]
To unsubscribe from this group, send email to
[email protected]
For more options, visit this group at
http://groups.google.com/group/otb-users?hl=en
--- 
You received this message because you are subscribed to the Google Groups 
"otb-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
For more options, visit https://groups.google.com/d/optout.

Reply via email to